A Multiphase Adjoint Model for CMAQ

Transcription

1 A Multiphase Adjoint Model for CMAQ Shunliu Zhao, Amir Hakami (Carleton University); Matt D. Turner, Shannon L. Capps, and Daven K. Henze (University of Colorado); Peter B. Percell (University of Houston); Jaroslav Resler (ICS Prague); Jesse O. Bash, Sergey L. Napelenok (USEPA);, Rob W. Pinder; Armistead G. Russell and Athanasios Nenes (Georgia Tech); Jaemeen Baek, Greg R. Carmichael, and Charlie O. Stanier (University of Iowa); Adrian Sandu (Virginia Tech); Tianfeng Chai (University of Maryland); Daewon Byun (NOAA) CMAS 2015

2 Outline Motivation for developing a multiphase adjoint model The current status of the development Adjoint code Process-by-process validation Test run of the full adjoint model Concluding remarks

3 Motivation CMAQ: Evolution of atmospheric gas and aerosol species Forward sensitivity analysis (Decoupled Direct Method, DDM) source receptor Forward Backward Backward/Adjoint sensitivity analysis

4 Adjoint sensitivity analysis: An example Pappin et al., Compounding Benefits of Air Pollution Control: A Revised View of Air Pollution Economics, Wednesday Presentation

5 The current status of the adjoint development CMAQ scientific processes: chemistry, aerosol, transport, cloud Adjoint code generated by tools (KPP, Tapenade, TAMC) or by hand Code validating using Finite Difference Method (FDM) Complex Variable Method (CVM) and sometimes DDM/TLM

6 Process-by-process validation Process Sub-processes Validation Methods Chemistry Aerosol Transport Secondary organic aerosol, Heterogeneous chemistry, Coagulation, Nucleation, Condensation, Thermodynamics (Isorropia; Capps et al., 2012) Horizontal/vertical advection, Horizontal/ vertical diffusion Cloud Convective cloud, Resolved cloud, Aqueous chemistry Tangent linear model

7 Process-by-process validation Process Sub-processes Validation Methods Chemistry Aerosol Transport Secondary organic aerosol, Heterogeneous chemistry, Coagulation, Nucleation, Condensation, Thermodynamics (Isorropia; Capps et al., 2012) Horizontal/vertical advection, Horizontal/ vertical diffusion Cloud Convective cloud, Resolved cloud, Aqueous chemistry Tangent linear model

8 Th adjoint of chemistry: Limitations of the FDM Adjoint

9 The adjoint of chemistry Adjoint

10 The adjoint of chemistry: Full Jacobian Adjoint

11 Process-by-process validation Process Sub-processes Validation Methods Chemistry Aerosol Transport Secondary organic aerosol, Heterogeneous chemistry, Coagulation, Nucleation, Condensation, Thermodynamics (Isorropia; Capps et al., 2012) Horizontal/vertical advection, Horizontal/ vertical diffusion Cloud Convective cloud, Resolved cloud, Aqueous chemistry Tangent linear model

12 The adjoint of aerosol: secondary organic aerosol Adjoint

13 The adjoint of aerosol: heterogeneous chemistry

14 The adjoint of aerosol: coagulation

15 The adjoint of aerosol dynamics Adjoint

16 The adjoint of aerosol: Isorropia

17 The adjoint of aerosol: aerosol dynamics and thermodynamics Adjoint

18 The adjoint of aerosol: aerosol dynamics and thermodynamics Adjoint

19 Process-by-process validation Process Sub-processes Validation Methods Chemistry Aerosol Transport Secondary organic aerosol, Heterogeneous chemistry, Coagulation, Nucleation, Condensation, Thermodynamics (Isorropia; Capps et al., 2012) Horizontal/vertical advection, Horizontal/ vertical diffusion Cloud Convective cloud, Resolved cloud, Aqueous chemistry Tangent linear model

20 The adjoint of transport: horizontal advection Horizontal advection at the X direction / discrete adjoint Adjoint Finite Difference

21 The adjoint of transport: vertical diffusion Adjoint Finite Difference

22 Adjoint The adjoint of vertical diffusion and chemistry 0.1 ppbv 1 ppbv 10 ppbv Finite Difference

23 Process-by-process validation Process Sub-processes Validation Methods Chemistry Aerosol Transport Secondary organic aerosol, Heterogeneous chemistry, Coagulation, Nucleation, Condensation, Thermodynamics (Isorropia; Capps et al., 2012) Horizontal/vertical advection, Horizontal/vertical diffusion Cloud Convective cloud, Resolved cloud, Aqueous chemistry Tangent linear model

24 The adjoint of clouds Aqueous chemistry (KPP; Kathleen Fahey, EPA) Adjoint DDM

25 Test run of the full adjoint model Day 1 Day 4

26 Test run of the full adjoint model Day 1 Day 4 without AQCHEM

27 Lessons learned Automatic differentiation entails a number of problems Significant clean-up debugging is necessary Single/double precision Active variables passed by modules not visible to some top routines Uninitialized variables in the adjoint code Problems that can be ignored in process-by-process tests may become more serious in interaction with other processes A number of issue can be avoided if forward model development is mindful of differentiation Fractured response surface for a number of processes

28 Concluding remarks When testing with the full adjoint model, blow-ups in the adjoint sensitivities caused by various reasons have been observed. Despite numerous bug fixes, large numbers still exist. Process-by-process validations are positive in general. But there seems to be interactions between processes which cause the abnormal growth in sensitivities.

29 Acknowledgements Funding: API, NSERC Model support: USEPA